Interested in how large language models compare against clinicians? Listen as Editor-in-Chief Steven George, PT, PhD, FAPTA, talks with Yael Arbel, PT, MHA, Yoav Gimmon, PT, PhD, and Liora Shmueli, PhD, about their recently published study, which tested the performance of ChatGPT and Google Gemini against 30 physical therapists and 30 physical therapist students on 20 multiple-choice questions about vestibular rehabilitation. The authors discuss the impact of prompt engineering, the potential for studies with other large language models, and the importance of human expertise when integrating new technological tools.
Arbel, Gimmon, and Shmueli are co-authors of the article "Evaluating the Potential of Large Language Models for Vestibular Rehabilitation Education: A Comparison of ChatGPT, Google Gemini, and Clinicians.” Read the article on the PTJ website and follow PTJ on LinkedIn and Bluesky for more research updates.